Upload
others
View
2
Download
0
Embed Size (px)
Citation preview
COMMONWEALTH OF AUSTRALIA
DEPARTMENT OF NATIONAL DEVELOPMENT
BUREAU OF MINERAL RESOURCESGEOLOGY AND GEOPHYSICS
RECORDS:
1967/17
TEE ALICE SPRINGS TOWN BASIN A CASE HISTORY
by
T. Quinlan
Prepared for the 1967 Groundwater School, Adelaide.'
The information contained in this report has been obtained by theDepartment of National Development, as part of the policy of the Common-wealth Government, to assist in the exploration and development of mineralresources. It may not be published in any form or used in a companyprospectus without the permission in writing of the Director, Bureau ofMineral Resources, Geology and Geophysics.
5928/65
THE ALICE SPRINGS TOWN BASIN
A CASE HISTORY
by
T. QUINLAN
RECORDS 1967/17
Prepared for the
1967 Groundwater School, Adelaide
The information contained in this report has been obtained by theDepartment of National Development, as part of the policy of the CommonwealthGovernment, to assist in the exploration and development of mdneral . resources.It may not be published in any form or used in a company prospectus withoutthepermission in writing of the Director, Bureau of Mineral Resources, Geology andGeophysics.
TABLE OF CONTENTS
‘Page No
SUMMARY
INTRODUCTION
History of Investigations
GEOLOGY
GEOPHYSICAL INVESUGATIONS
11
3
4
HYDROLOGY^ 5Availability of Groundwater^ 5Methods of Construction and Development of Bores^ 5Run-off and River Flow^ 7Chemical character of Groundwater^ 8
Recharge^ 9Water Levels^ 10
Aquifer Performance Tests^ 11
Volume of Saturated Alluvium^ 12
Safe -Yield^ 12
CONCLUSIONS^ 14
REFERENCES^ 15
TABLES
1. SPeditic capacities of bores
2. Coefficients calculated from aquifer performance tests
3.^Storage and withdrawal of groundwater.
•
PLATES
1. Geological map of the Town and Inner Farm Basins Alice Springs.
2. Geology and salinity of groundwater,, Quaternary aquifers, Town Basin,Alice Springs.
3.^Chemical character (anions) of groundwater, Quaternary aquifers, TownBasin, Alice Springs.
• Fluctuations in the piezometric surface, Town Basin, Alice Springs.
5.^Drawdown curves for aquifer performance tests, Town Basin, Alice Springs.
FIGURES
1. Locality map and regional geological map of the catchments of the Todd and'Charles Rivers.
2. • Lithological log and completion details of bores 62/8 and 62/9.
3. Consumption and withdrawal of groundwater.
4. Calculation of safe yield.
SUMMARY
The Alice Springs Town Basin is a small basin which has been filled to
a maximum depth of 75 feet with alluvial sediment of Quaternary age. Severalmethods have been tried to construct efficient bores which could be pumped for the
town supply. It contains groundwater of four different chemical types; ground-
water suitable for domestic use occurs in the vicinity of the Todd River. The
volume of water in storage fell from 647 million gallons in 1953 to 357 million
gallons in 1964, as a result of an increase in the amount of water pumped from
the basin. Average values for the aquifer constants (30,000 gallons per day per
foot for the coefficient of transmissibility and 0.07 for the coefficient of
storage) were calculated from aquifer performance tests.
Maximum use could be made of the water available as recharge if ground-
water was withdrawn at a rate of 20 million gallons per month for a period'of 2
months following a river flow, and then reducing the withdrawal rate by 3.5million gallons per month until the next period of recharge.
Oa Alluvium, river grovel, colluvium
Road
^ -- Vehicle track
4--1-•-s--1- Railway
t Landing ground
^
A^Trigonometric station
•1898'^Height in feet, barometric (datum-mean sea level)
Reference
Sandstone, conglomeratic sandstone, calcareous siltyPiir..*:•:::11 sandstone, limestone, trovertine, conglomerate, chalcedony
Dolomite, limestone, siltstone, sandstone, gypsum
r 'e;h‘.."1 ouorrrire
•Granite
Mica - quartz -feldspar schist and gneiss ,garnet- 'mica- feldsparpf c,...1 gneiss ,quortzo-feldspothic gneiss, amphibolite, meta- brialc
rocks, meto-limestone, quartzite, pegmatite • granite
Zr
Arunta Complex
Grid: 10,000yd military grid for zone 5
5 MILES
^ Geological boundary'
Fault
Trend lines
Strike ' and dip of foliation
Bore with windpump
• •^Homestead
18
Bitter Springs Formation
Heavitree Quartzite
To accompany Records 1967/17
Fig. I Locality map and regional geological map of the catchments of theTodd and CharlesRivers
F53/A/4/60
INTRODUCTION
The Alice Springs Town Basin is a small alluvial basin with A maximumdepth of approximately 75 feet and a surface area of approximately 3 squaremiles. Groundwater is withdrawn from the basin far the Alice Springs town supply,
and until January 1964 this was the only source of supply.. Water for the town
supply was imported from the Inner Farm Basin in limited quantities from January
1964 to April 1965. In the latter part of 1964, pumping from the Town Basin
virtually ceased when production from an alternative source ofgroundwater in
Palaeozoic sandstone was commenced. The annual consumption at that date wasabout
250 million gallons, of which approximately 200 million gallons was withdrawn bythe town supply bores.
Between 1943 and 1964 information on the occurrence of groundwater and the
hydrology of the basin was collected by residents of the town, the Australian Army,
the Commonwealth Department of Works, the Water Resources and Animal Industry
Branches and the Resident Geologists of the Northern Territory Administration.
. Some aspects of the methods used to collect this infOrmation and the inter-
pretations which were placed on it are discussed in this report.
History of Investigations
Test drilling of the Town Basin (Pl. 1) was begun by the Australian Army
in 1943 and was continued by the Department of Works from 1953 to 1961.. Theearliest reports are those of Owen (1952, 1954), who considered salinity, recharge,
and groundwater movement. ,He used a bedrock contour map to arrive at. an estimate
of 900 million gallons for the amount of water stored in the basin in 1954.
Jones (1957) was able to refine Owen's ideas and to prepare a more accurate
contour map, with the aid of the additional information available. He estimated
the amount of water stored in the basin to be 330 million gallons.
-2-
The Bureau of Mineral Resources carried out resistivity and seismic
surveys in the ToWn and Inner Farm Basins in 1956 (Dyson & Wiebenga, 1957). The
results of -these surVeys were used.in the planning of part of the Department of
Works investigatory drilling progratme.
Wilson (1958) conducted an investigation for the 'Department of Works
and examined groundwater movement, salinity distrihution, and groundwater storage.
He estimated that 1110 million gallons of water were stored in the Town Basin.
He also carried out pumping tests on some of the production wells and measured .
some flows of the Todd River.
Forbes (1962) studied groundwater movement and salinity in both Town and
Inner Farm Basins and estimated that the annual safe yield from the Town Basin
Would be 149 million gallons. He also estimated the average annual yield of
surface run-off at Heavitree Gap to be 3180 million gallons.
Jephcott (1959) studied the relationship between groundwater salinity and
river flow in the Town Basin, and concluded that sodium and bicarbonate are the
most sensitive indicators of basin recharge.
Quinlan & Woolley (1962) discussed the occurrence of groundwater in the
Town Basine using the results of the drilling programme to 1960. They drew
attention to the decline in the volume of groundwater in storage from 1953 to 1960,
and made a preliminary interpretation of the results of some of the pumping tests.
Subsequently they (Quinlan & Woolley, 1967) refined their concept of the geology of
the basin:and presented it in maps of five aquifer systems. These were used as
abase to illustrate the variation in the chemical character of the groundwater.
Further estimates for hydrological parameters weregiven and an elementary attempt
was made to predict the change in storage which would result from three methods of
management.
- 3-
GEOLOGY
Fluviatile sediment assigned to the Quaternary period is preserved in the
Town Basin, and rests unconformably on .Precambrian sedimentary, igneoUs and
metamorphic rocks. These facts together with the approximate position of its
margins were known to Owen (1952). However little was known of —the...'distribution
of the different grades of material of which the beds are composed, and it is
improbable that the sediments of the basin, having been laid down and reworked by
a meandering stream, contain any particular bed that persists for any great
distance.'
Jones (1957) recognized the importance of such knowledge, as the available
evidence indicated that 'less than half the alluvium yields water readily'.
Sufficient information became available as a result of the programme of test
drilling undertaken by the Department of Works in 1956 and 1957 to form an initialhypothesis concerning the distribution of aquifers. This was not dime at the time
as a geologist was not associated with the project. At that time adequate logs were
available for approximately 150 bores.
Quinlan and Woolley (1962), using the results of drilling to 1961 .postulated
a sequence of events during the deposition of alluvium in the basin. This led to
the concept of linear zones of high permeability and the presence of hydraulic
boundaries within the alluvium. That this concept was somewhat idealistic is
evident from a comparison with later work (Quinlan & Woolley, 1967). The geological
maps of the five aquifers (Pl. 2) prepared by them were based on subjective
correlations between bore logs. This might not be a.satisfactory basis for mapping,
but it is justified as some understanding of the geology of the basin,, and the *
distribution of aquifers within it, is aprequisite for the development of
hYpotheses regarding the 'hydraulic behaviour of the basin, and the chemical
character of groundwater which it contains.
-4-GEOPHYSICAL INVESTIGATIONS
Seismic and resistivity surveys were undertaken by Dyson & Wiebenga
( 19 57 )Two maps of the areal distribution of their apparent resistiVity-measure-
ments were prepared, one for an electrode spacing of 200 feet and the , otherfor a
spacing of'400'feet. The apparent resistance for each is thought to be 'largely
influenced by the resistance of the Precambrian rocks and only prominent geol-
ogical and hydrological features are represented in the contour maps.
Quinlan and Woolley (1967) attempted to estimate the porosity of the
alluvium by using the monogram of Dyson & Wiebenga (1957, fig. Al). Values of
the resistance of the saturated alluvium obtained from the depth probes were
compared with values of the salinity of the pore fluid, estimated from the
analyses of water samples taken from bores in the vicinity. Most of the estimates
for the porosity were found to be greater than 30 percent and are not considered
to be realistic.
Using the seismic refraction method, Dyson & Wiebenga (1957) distinguished
two digeOritihuities which they considered to be (a) the top of the zoner4i±:
saturation, and (b) the base of the weathered zone in the Precambrian-rodks.
The,base of the zone of weathering is irregular and its thickness is,
influenced by the lithology of the parent rock and its susceptibility to weathering.
Assuming that these factors are constant, the base of the weathered zone will be at
about the same depth below the surface of the Precambrian. Having made this
assumption the contoured map prepared from the results of the seismic refraction
survey was valuable for planning drilling programmes. Its usefulness decreased as
the number of bores increased and contours could be drawn on the surface of the
Precambrian with some confidence....^ ,
-5-
HYDROLOGY
Availability_of Groundwater
Groundwater can be extracted from bores and wells which intersect
permeable beds of silty sand below the piezometric surface. The specific
capacities of bores within the areas of silty sand which are mapped on Plate 2 may
be over 1000 gallons per hour per foot of drawdown. The capacities of bores
outside these areas is variable, and will probably be less than 3000 gallons per
hour per foot of drawdown.
Until 1956 wells were used to withdraw water for the town supply and for
private.consumption from shallow aquifers. These proved to be unreliable in the
succeeding years because of a general decline in the piezometric surface; and they
were replaced . by bores which were able to withdraw water from the deeper aquifers.
The fall in the pie zometric surface was sufficient to cause a significant
reduction in the area of the saturated alluvium and volume of groundwater available.
Methods of Construction and Development of Bores ,
Included within the funds made available in 1956 for the replacement of
the town supply wells was a provision for test drilling. The test. holes, were
designed to obtain information on the rapid lateral and vertical changes-in
lithology, to evaluate the resources of the basin, and to find suitable sites for
the construction of production bores.
Wilson .(1957) drilled approximately 90 bores, of which one was completed
with screens and five with perforated casing as production bores. Three of the
latter were unstable and were subsequently abandoned. Quinlan and Woolley were
responsible for drilling approximately 100. bores between 1959 and 1962, of which 8were completed as production bores.
Little is known of the methods used by Wilson, and the following
discussion is based on Quinlan & Woolley (1967). Contractors with percussion rigs
were engaged to drill under the supervision of a geologist. The holes were drilled
-6-
with a chisel bit and without casing to the first aquifer. A'string of 6-inch
blank casing was then run into the hole, and casing measurements were kept to
check on the position of the casing shoe in relation to the bottom of the hole.
Below the piezometric surface the casing shoe was driven ahead of the hole, except
when in thick intervals of clay or silt where an open hole was drilledahead.of the
shoe for:2 Or- .3 feet to.free the casing.
When drillingclaystand silts the hole had to be cleaned regularly with
the sand pump before an uncontaminated sample could be obtained and water was added
to the hole to minimize the risk of sand entering the hole around the casing shoe
together with water from aquifers behind the blank casing. This practice made it
difficult to obtain a true water sample from an aquifer, and to estimate the yield,
but these precautions were necessary to obtain an accurate log of the hole..,
Most of the production holes were completed with sandscreens'swedged inside
8-inch casing. In the earlier production holes, a screen-slot opening was selected
which would allow about 50 percent of the sand near the screen to pass through.
Two lengths of screen were generally placed opposite the deepest and thickest
aquifer located by test drilling. The,holes were developed by means of mechanical
surging, using both solid and valve surge tools. Sloughing of silt and clay from
the walls was a frequent occurrence, because in most of the holes there are several
thin aquifers separated by silt or silty clay, and it was impracticable to-place
screen opposite each aquifer. Eight inch casing was necessary to accOmmodate the
pumping equipment.
Attempts were made to prevent collapse by using gravel packing techniques,
but these were not successful in stabilizing all the aquifers in a particular hole.
Where separate gravel feed holes were used. it was:not,possible to control-.the
emplacement of the gravel with sufficient accuracy. An attempt to provide a gravel
envelope in the annulus between the production casing and the drill casing failed,
probably- because it was impracticable to use casing large enough to provide a.
sufficiently wide annulus.
TABLE 1.^SPECIFIC CAPACITIES OF BORES
Bore^Method of^Amount of Sand^Specific Construction^Passing through^Capacity
Screen
(%)^
(gph/ft drawdown)
Well Loss
(% of drawdownat 3 0 0004Ph)
430 29
120 20
270 10
1. Perforated casing^80-90
2. Casing withdrawn to^50expose screen
80
-^ 300^ 5
85^ 450^ 14
6.^Screen undercut^80^ 1000 •^14into place
3. u
4. 1,
5. 11
. •, x6 ,5,00,6- !).Wills screen
21-25 Grey silt and cloy
62/9 62/8Protector^R L 159235'welded toCasing
Natural Surface Cot. /962
Brown, medium to
coarse-groined silty
sand
AQUIFER Ab 1
7 - /5 Brown, fine-grained
sandy silt and cloy
15 - /7 Brown and grey, fine to
very coorse-groined ckyey
sand and gravel
/7 -19 Bros re,coorse to very coarse-
mined soy sand
19 - 21 Grey,very COOrse -9/01Red
ckyey sand
AQUIFER 602
SW L Dec 62 -
4 8"X 0072"Wills screen-`""-
AQUIFER 1103
25-27 Grey and brrem, worse to
very coorse-grained clayey sand
27-29 8,-con, fine to very coarse-
grained sand and grovel
29 - 38 Grey cloy and
46-50 Grey and Omen, fireet-grakeedPacker
sandy cloy and silt
Scale0^5 FEET
occompaty Records /967/17
ect,
ec)
38 - 46 Grey and Drown, fine to
very coarse-grained
clayey sand
Fig.2. Lithological log and completiondetails of bores 62/8 and 62/9
-7-
In some of the later bores, screens with larger slot openings were used
and development was restricted to pumping only, to avoid collapse of aquifers.
This was only partly successful, although it was quite successful in developing
the screened aquifer.
As the annular space between the wall of the hole and the screen may be
sufficient to start collapse, two methods were tried in which the openings in the
production string were set opposite the aquifer without withdrawing the casing.
In the fist method 8-inch casing with 4-inch holes drilled on a 14 inch square
grid was used. The success of this method is probably due to production from two
or more aquifers. However the well loss in such a bore is considerably higher than
in bores completed with sand screens (see Table 1).
The second method was to undercut into place a production string of screens
screwed to 8- inch casing (Fig. 2). A 20 foot length of blank casing,was necessaryat the bottom of the string to allow the undercut bit to work properly.., Development
was by pumping with an axial flow turbine pump, with backwadhing by stopping the
pump. With this method it was possible to construct a bore with, a specific
capacity of approximately 1,000 gallons per hour and a low well loss.-
The specific capacity and well loss as a percentage of total drawdown at a
pumping rate of 3,000 gallons per hour, of the different types of bores are listed
in Table 1.
Run-off and River Flow
The amount of rainfall required on catchment before run-off will begin is
variable and depends on the intensity of the rain and the previous history, of .
qualitative estimates of this amount have been made: Wilsor,0958)
considers. that 'the Todd River appears to flow only after about 1' of rein'of an
intensitapp±bximating to 2 1' per hour', and Forbes (1962) concluded that'en"
average rain of 40-50 points of sufficient intensity in one day will bring . theriver down.
-8-
These figures, and the daily rainfall totals and recorded volumes of river
floWs„ have been manipulated to estimate the yield of surface run-off from the
catchments of the Todd and Charles Rivers. Wilson (1958) considered that an
average of three medium flows crossed the East Side Causeway, during the period
1937-57, with total average yield of 600 million gallons per year. Forbes (1962)
extrapolated a relation between average annual rainfall and surface flow through
Heavitree Gap to derive an average annual yield of 507 million.gallons.
A more satisfactory method for estimating the annual yield would be touse the frequency distribution of discharge, estimated from stream gauging,
measurements, when these become available.
Chemical Character of the Groundwater
The results of water analyses performed in the chemical laboratory of the
Animal Industry Branch, Northern Territory Administration, Alice Springs, have been
the basis for all interpretations of the chemical character of groundwater.
Jones (1957) Constructed a contoured map of the content of total dissolved
solids (Tps) of the water in the basin, and demonstrated that the better qualitywater occurred in the vicinity of the Todd River.
Quinlan & Woolley (1967), using additional information, constructed TDS
contours for each of their five aquifer systems (Plate 2). These maps'ehOw that
there . is a relation between the salinity of the water, the Variation in pertea-
bility.of the aquifer, and the distance to a source of recharge water.
Forbes (1962) , used the graphical method of Schoeller (1959) to distinguish
8 chemical. types of groundwater. He concluded that this method 'was of little valuefor this basin.
Quinlan & Woolley (1967) used a method based on the traditional triangular
plots Of a three component end member system. The cation and anion content were
considered separately, and the position of the point plotted for a wateranalysis
was described by the classifying function °D° of Pelto (1954). A coherent but
subjective interpretation of the areal variation in chemical character was obtained
by contouring these values of 'DI, within areas in which one or more anions were
dominant constituents (Plate 3). They found that sodium was the dominant cation
throughout the basin, but the analyses failed to show a systematio variation for
the cation content. They recognized that changes in the hydraulio regime during the
period 1954-64 were responsible for some changes in the areal variation but these
had to be ignored because of the lack of data.
The data which are available does provide some information on the change
with time of the ohemioal character of water in some areas. It is assumed that the
addition of recharge water from the Todd River lowers the salinity of the ground-
water and changes its composition until equilibrium has been re-established, and
the diffusion of ions from the groundwatei -to the recharge water is an important
mechanism in the process. The rates of diffusion are low, and a considerable time
may ellapse between the start of river flow and the resultant changes in composition
of the groundwater at some distance from the river. The changes can be expected to
decrease with the distance from the river.
Jephoott (1958) concluded that the most sensitive indicators of the movement
of recharge water through the basin are the sodium and bicarbonate ions. It would
appear that the concentration of the bicarbonate, sodium, and potassium ions is less
in floodwater than in groundwater whioh is in the vicinity of the river (Quinlan &
Woolley, 1967). The bicarbonate or carbonate ion is not normally a component of the
minerals in alluvium, and the most probable souroe appears to be the carbon dioxide
in the air trapped in the pores of the alluvium. The additional sodium and potassium
ions have presumably been derived from the minerals in the aquifer.
Recharge
Recharge to all aquifers originates as floodwater in the Todd River. Certain
portions of the river bed are known to be recharge areas whioh are interconnected
with aquifers within the basin. ObserVaions on the rate of advance and recession
of floodwaters in the recharge areas indicate that the rate of infiltration of water
is signifioantly greater than in other parts of the channel. Wilson (1958) records
initial infiltration rates for the Todd River as high as 180 gallons per hour per
square foot, within 24 hours this rate had dropped to 0.5 gallons per hour per square
foot. Factors which would inhibit infiltration are the swelling of the clay matrix
-10-
following Wetting and the presence of air trapped in the pore space. . It is doubtful
if the river ever continues to run for a sufficient period to dissipate the
entrapped air.
The recharge Mound in the piezometric surface adjacent to the:Todd River
encloses permeable allUvium,in which recharge water is stored under unconfined
conditions until it moves into confined aquifers.
Water Levels
Water level hydrographs for representative bores in each of the five
Quaternary aquifers are shown on Plate 4. They show that the water level fluctuates
after floods in the Todd River and in response to pumping near the observation bores.
The magnitude and character of the fluctuation depends on the distance , of the,
observation bore from, the recharge areas, the crests of the hydrographe-becOme
rounded, and the rate of decline in water levels is much less.
The rise in water levels in response to a river flow which occurs within
3 months of a previous flow is generally not as great as the rise restating from
flows separated by a longer interval. A minimum period of 3 months is probablyrequired-for - the recharge mound near the river to be sufficiently dissipated to
permit significant quantities to infiltrate -fromsubsequent flows.
Water' levels in the basin fell during the period 1958-64, aea result of
the increased withdrawal in the same period (Pl. 4 and Fig. 3). The.rate of decline
was greatest -in the south and was not significant in the north. The decline hae
not apparently induced'recharge to the basin as predicted by Wilson (1958) .; The
water levels in the bores in the areas of greatest decline (J and 59/8) showed a
much smaller response' to river flow in 1962 than in Previous years., Yet : the river
ran seven times in 1962. It follows that the rate" of transfer of water from the
recharge mound to the five aquifer systems must be less than the rate of movement
of groundwater within the aquifers.
10- Number of To. $upply .115 ona bores
^
5- 1̂
^
300 ^
0..ity of water withdrawn horn Town Basin250-
in previous 12 months (million son)
200-
150-^ 4^•
100 .'
50-
0
1953^954^1955 . 956^1951^1958^959 , 1960^1961 , 962 , 1965 , 964
Population of Alice SPd.11 5
6.000
4,000-
100.
2. 000-
1, COO.
200.Dads consumption from Tow, 5.001Y (941100snst.0 )
600* rte., 445 TOL 1000 ppm
5001953^1954^955^1956^M57 • 19513^1959 ' 1960^1961^1962^963^1964
1,400
1,30. -
1. 100-,
1.005
Fig.4. Calculation of safe yield
p.
Fig.3. Consumption and withdrawal ogroundwater
To accompaoy Records 196'7//7
F53/A/4/62
TABLE 2,
Bore^Non-Equilibrium.Method
Coefficients Calculated from Aquifer Performance Tests
EquilibriumMethod
Coefficient ofTransmissibilityin imperial gallonsday/foot
HydraulicConductivityin feet/day
Coefficientof Storage
Coefficient of ,
Transmissibilityin gallons/day/ft.
HydraulicConduc-tivity infeet/day
Coefficient ofStorage
62/9 4750 log 0.05
Colocag Park 26,000 95 0.08
61/24
Bent Tree No. 1
60/14 38,000 187 0.13
Todd Bore 60/5 31,000 125 0.09
61/33 35,500 200 0.004
110 22,000 111 0.03 56,000 280 0.00014
Army Well No. 2 17,000-66,000 150 0.07 30,000 190 0.43
59 42,000 120 0.02
28 20,000 93 0.07
59/11 5,000-42,000 130 0.002 17,000 0.0001-0.0001
Town Wells recal-culated from
( Wilson 1957)
Bent Tree Well
calculated from
12,000 92 0.17
Wilson (1957)
Values adopted are thought to be applicable to the Town Basin: Coefficient of Transmissibility . 30,000 gallons/day/foot
Hydraulic conductivity^= 150 feet/day
Coefficient of Storage^. 0.07
Aquifer Performance Tests
Wilson (1958) and 'Qu. an & Woolley (1962, 1967) carried out a number ofaquifer performance tests on t . -is and wells. The former were of one or two days
duration while the latter were 1:un for as long as three months.
QUinlan and Woolley:ud bores which were pumping water for the townsupply to avoid the installatun of special pumps. Several factors'increase the
probable error of their result2:
1. Tests were possible only during the winter months, when adjacent pro-
duction bores could be shut down to avoid interference between bores. The time
available for'testing was seldom long enough to measure the recovery on- completion
of pumping. .-
2. 'The discharge may fluctuate by percent, because of variations in headin the 'riding mains.
3.^. The. observed drawdowns were adjusted to allow for the natural: fall in
water levels between recharge periods. These corrections were obtained by extra
polationifrot long-term trends in water levels at observation points, both inside
and outside the area of influence.
Because of the irregular distribution and thickness of the channel sands
it is not': pOseible to treat them individually as confined aquifers, or, More
properly as infinite-strip leaky artesian aquifers, and a. choice of an appropriate
model must be made to interpretate the results.
Quinlan and Woolley assumed. that on a long term basis, the withdrawal of
water from the saturated alluvium can be considered to occur under unconfined
conditions, and that it was expedient to assume that the alluvium is isotropic,
homogeneous and of infinite areal extent. In which case the consistency.of theresults from- one test can be taken as an indication of the differences between the
ideal aquifer and the saturated alluvium. Both the Theis non-equilibrium formula
(Wenzel, )942) and the equilibrium formula (Jacob, 1950) were used (Plate 5) and
the results are given in Table 2. The range in values calcUlated from drawdowns
-1 2-
in several observation holes for a particular test was large, and tan:be taken as an.
indication of the poor agreement between the model and the saturated alluvium. The
average values for the hydraulic ,conductivity for eadhtest.-ite more-consistant, and
this is considered to justify the application of one model to the whole basin.
Aolume of Saturated Alluvium
A graphical method was used to estimate the volume of saturated alluvium in
the basin from contoured maps of the piezometric surface. It was calculated for a
date in October for each year from 1953-64 inclusive, and for April in 1963 and 1964.
Estimates of the volume of groundwater stored in the basin (Table 3) were
based on the calculated volume of saturated alluvium and a specific yield of 0.07
(Table 2). They range from 357 million gallons in October 1964 to 649 milliongallons in October 1955, compared with the 900 million gallons estimated by Owen
(1952, 1954). 1,110 million gallons by Wilson (1958), and 300 million gallons by
Jones .(1957).
Safe Yield
It must be stressed that while it is desirable to estimate a Safe yield to
allow for reasonable planning for development, such an estimate can, in Many cases,
only be made without an adequate appreciation of the hydraulic behaviour of the basin.
The eetimates which have been made to date are based on data collected during aperiod of depletion. Until the water levels have recovered to a state comparable : .with that existing before 1957, there is no basis for comparison to detertine if itshydraulic behaviour has been adversely affected by depletion.
Wilson (1958) considered that the yield of the basin could be maintained at
340 million gallons pr year, I providing that there were three average floodeinthe ,3
river and that the qu tity,of wtter received from recharge increased as thehydraulic gradient incr sed with depletion.
3TABLE/: STORAGE AND W1THDRAWAL.OP GROUNDWATER, ALICE SPRINGS TOM BASIN
Year Ending Volume ofSaturated : "AlluviuM
(10° cu. ft.)
Estimated Volumeof Groundwaterin Storage (million gals
,'for.
Specific yield . 7%),
Estimated Volumeof GroundwaterPumped from Basin(million gals)
Change in Volumeof SaturatedAlluvium(106 cu. ft.)
EstimatedPeriodwithoutRecharge(days)
PredictedSafe Yield(million• -gals)
Calculated Predicted -Difference
1st Sept. 1953 1480 647.5 so 264October 1954 1 392 609.0 120 -88 -94 6 236 43October 1955 1483 648.8 130 +91 117 -26 70 278October 1956 1440 630.0 141 -43 -44 146 17616th Oct. 1957 1349 590. 2 152 - -91 -75 -16 192 107October 1958 1420 621.2 160 +72 +35 +36 103 233October 1959 1213 •530.7 175 -208 -20 77 265 14October 1960 1074 469.9 192 -139 -182 43 235 5514th Oct. 1961 1167 510.6 206 +93 +15 +78 76 26911th Oct. 1962 1065 465.9 256 -102 -8o -22 101 23618th Oct. 1963 980 428.7 276 -82 -57 -25 66 2832nd October 1964 817 357.4 ' 219 -166 -143 -23 182 126
TOTAL
-665 -67o
-13-
Forbes (1962) used the Hills method (Todd, 1959) to obtain an estimate of
120 million gallons per year. This was based on the position of the zero change in
water level on the straight line of best fit to a plot of the quantity of water
withdrawn from the basin and the rate of decline of water levels in bore HC. He
recognized the limitations of the method as applied to small basins. These are
significant in this case, as the bore HC is close to the western margin of the
saturated alluvium.
Quinlan and Woolley (1967) adapted the method by using the annual change
in volume of the saturated alluvium instead of the change in water level. In
addition they included a second variable, the estimated number of days in each year
when there was no recharge, on the assumption that recharge will continue for 100
days, which is the average duration of the rise in water levels after a river flow
(pl. 4). Using an analysis of multiple regression they obtained the relation
y = 372' - 1.23x . -^1.35t,
where y is the change in the volume of saturated alluvium, x is the amount of water
pumped from the basin in the year, and .t is the estimated number of'dayesduring the
year without recharge. The basic assumption that a linear relatiOn exists between
the variables is probably not realistic, and the relation was considered to be an
approximation.
The value obtained for the multiple correlation coefficient was relatively
high (0.89) and there was reasonable agreement between the total of the ; predictedchange (668 million cubic feet) and the observed change (665 million cubic feet) forthe years 1953 to 1964. Both these facts indicate that this equation would bereasonably effective in predicting the changes over long periods. The differences
between the.observed,and the predicted changes in volume for each.year.indicate the
limitations of the method for the prediction of Short term changed.
They believed that a single value cannot be assigned to the annual safe
yield, as it is largely dependant on the number of river flows, separated by more
than 100 days,' which have occurred during the year. The safe yields which can be
expected are illustrated in Table 2 and Figure 4. On this basis there are threepossible approaches to the management of the basin.
-14-
1. Within a period of 100 years, the Todd River will flow at least once
during-91 of the years. If this is the level of probability at which the basin is
managed, then a value for the annual safe yield is insignificant. This is obviously
impracticable as a basis for management,
2. Management at the 50 percent level of probability (i.e. depletion would
occur in 50 years of a 100 year period, and water would be added to storage during
the other 50 years) would allow the withdrawal of groundwater at a rate of 156
million gallons per year. This estimate is based on the probability distribution of
the number of days per annum when recharge does not occur, for which the median is
126 days.
3.^The third and recommended method would take advantage of each recharge
period as it occurred. This could be achieved by the withdrawal of groundwater at
a rate of 20 Million gallons per month for a period of 2 months following recharge,
'then reducing the withdrawal rate by 3 ..5 million gallons in each succeeding month,until the next period of recharge. The probability of failure of such:..a policy was
not given 'but.it ‘ could be expected to be low.
CONCLUSIONS
Much information concerning the geology of the Alice Apringp Town Basin
and on the occurrence of groundwater in it, has beam* available as a result of the
work undertaken by a number of organizations during the period 1952-64.,
The analysis and interpretation of water level measurements and.pumping
test data has not provided a satisfactory model which could be used to describe
and predict its hydraulic behaviour. Further work is in progress to resolve this
problem.
- 15-
REFERENCES
DYSON, D.F. p and WiEBENGA, W.E., 1957 - Final report on geophysical investigationsof underground water ) Alice Springs N.T. Bur. Min. Resour. Aust. Rec.
1957/87 (unpubl.).
FORBES, C.F, 1962 - Estimation of safe yield from Alice Springs Groundwater Basin
N. Terr. Admin., Water Resour. Br., tech. Rep. 1962/9.
JACOB, C.E., 1950 - Flow of groundwater, in Engineering Hydraulics (H. Rouse,
Wiley, New York 321-386.
JEPHCOTT,-B.0959 7 The correlation between salinity and river flow in. the Alice
Springs town water supply. Roy. Soc. South Aust. Trans. 82.
J0NESO4 ..0. 1 1957, - Preliminary report on the ground water reserves : of - the Alice
.Springs area:. Bur. Min. Resaur. Aust. Rec. 1957/6 (unpub].).
OWEN, H.B„:1952 - Notes on water supply at Alice Springs. Bur. Min. Redour. Aust.
Rec. 1952/21 (unpubl.).
OWEN, H.B., 1954 - Notes on water supply. Alice Springs. ur. Min. Resour. Aust. Rec.
1952/21 (unpubl.).
PELTO, e.R., 1954 - Mapping of Multi Component systems, Jour. Geol., 62 (5) 501.
QUINLAN, T., and WOOLLEY, D.R. 1 1962 - The occurrence of groundwater in the AliceSprings Town Basin. Bur. Min. Resour. Aust. Rec. 1962/75 (unpubl.).
QUINLAN, T., and WOOLLEY, D.R. 2 1967 - Geology and hydrology of the Alice Springs.Town and Inner Farm Basins, Northern Territory. Bur. Min. Resour. Aust.
BUll. 89 (in press).
SCHOELLER, Hop 1959 - Arid Zone Hydrology. Recent Developments U.N.E.S.C.O. Paris.
-16-
TODD, D.K. 9 1959 - Groundwater Hydrology - John Wiley & Sons, Inc. New York.
WENZEL, L.K.; 1942 - Methods for determining permeability of water bearing
materials.^U.S. Geol. Sur. WS? 887.
WILSON, T., 1958- Report on engineering investigations of the water resources ofAlice Springsin the Northern Territory of Australia. Comm Dept. of Works
(unpubl).